An efficient approach for multi-view human action recognition based on bag-of-key-poses

Alexandros Andre Chaaraoui; Pau Climent-Perez; Francisco Florez Revuelta

doi:10.1007/978-3-642-34014-7_3

An efficient approach for multi-view human action recognition based on bag-of-key-poses

Alexandros Andre Chaaraoui
, Pau Climent-Perez
, Francisco Florez Revuelta

University of Alicante
Kingston University

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

Abstract

This paper presents a novel multi-view human action recognition approach based on a bag-of-key-poses. In the case of multi-view scenarios, it is especially difficult to perform accurate action recognition that still runs at an admissible recognition speed. The presented method aims to fill this gap by combining a silhouette-based pose representation with a simple, yet effective multi-view learning approach based on Model Fusion. Action classification is performed through efficient sequence matching and by the comparison of successive key poses which are evaluated on both feature similarity and match relevance. Experimentation on the MuHAVi dataset shows that the method outperforms currently available recognition rates and is exceptionally robust to actor-variance. Temporal evaluation confirms the method’s suitability for real-time recognition.

Original language	English
Title of host publication	Human behavior understanding
Subtitle of host publication	Third Workshop, HBU 2012, Vilamoura, Portugal, October 7, 2012, proceedings
Editors	Albert Ali Salah, Javier Ruiz-del-Solar, Çetin Meriçli, Pierre-Yves Oudeyer
Place of Publication	Berlin, Germany
Publisher	Springer
Pages	29-40
Number of pages	12
Volume	7559
ISBN (Electronic)	9783642340147
ISBN (Print)	9783642340130
DOIs	https://doi.org/10.1007/978-3-642-34014-7_3
Publication status	Published - Oct 2012
Externally published	Yes
Event	3rd International Workshop on Human Behavior Understanding - Vilamoura, Portugal Duration: 7 Oct 2012 → 7 Oct 2012

Publication series

Name	Lecture Notes in Computer Science
Publisher	Springer
Volume	7559
ISSN (Print)	0302-9743
ISSN (Electronic)	1611-3349

Conference

Conference	3rd International Workshop on Human Behavior Understanding
Period	7/10/12 → 7/10/12

Bibliographical note

Organising Body: Springer

Keywords

human action recognition
multi-view action recognition
key pose
bag-of-key-poses
MuHAVi dataset
Computer science and informatics

Access to Document

10.1007/978-3-642-34014-7_3

Cite this

Chaaraoui, A. A., Climent-Perez, P., & Florez Revuelta, F. (2012). An efficient approach for multi-view human action recognition based on bag-of-key-poses. In A. Ali Salah, J. Ruiz-del-Solar, Ç. Meriçli, & P.-Y. Oudeyer (Eds.), Human behavior understanding: Third Workshop, HBU 2012, Vilamoura, Portugal, October 7, 2012, proceedings (Vol. 7559, pp. 29-40). (Lecture Notes in Computer Science; Vol. 7559). Springer. https://doi.org/10.1007/978-3-642-34014-7_3

Chaaraoui, Alexandros Andre ; Climent-Perez, Pau ; Florez Revuelta, Francisco. / An efficient approach for multi-view human action recognition based on bag-of-key-poses. Human behavior understanding: Third Workshop, HBU 2012, Vilamoura, Portugal, October 7, 2012, proceedings. editor / Albert Ali Salah ; Javier Ruiz-del-Solar ; Çetin Meriçli ; Pierre-Yves Oudeyer. Vol. 7559 Berlin, Germany : Springer, 2012. pp. 29-40 (Lecture Notes in Computer Science).

@inproceedings{fe8bde538e174dd2815ab64b5f194dac,

title = "An efficient approach for multi-view human action recognition based on bag-of-key-poses",

abstract = "This paper presents a novel multi-view human action recognition approach based on a bag-of-key-poses. In the case of multi-view scenarios, it is especially difficult to perform accurate action recognition that still runs at an admissible recognition speed. The presented method aims to fill this gap by combining a silhouette-based pose representation with a simple, yet effective multi-view learning approach based on Model Fusion. Action classification is performed through efficient sequence matching and by the comparison of successive key poses which are evaluated on both feature similarity and match relevance. Experimentation on the MuHAVi dataset shows that the method outperforms currently available recognition rates and is exceptionally robust to actor-variance. Temporal evaluation confirms the method{\textquoteright}s suitability for real-time recognition.",

keywords = "human action recognition, multi-view action recognition, key pose, bag-of-key-poses, MuHAVi dataset, Computer science and informatics",

author = "Chaaraoui, \{Alexandros Andre\} and Pau Climent-Perez and \{Florez Revuelta\}, Francisco",

note = "Organising Body: Springer; 3rd International Workshop on Human Behavior Understanding ; Conference date: 07-10-2012 Through 07-10-2012",

year = "2012",

month = oct,

doi = "10.1007/978-3-642-34014-7\_3",

language = "English",

isbn = "9783642340130",

volume = "7559",

series = "Lecture Notes in Computer Science",

publisher = "Springer",

pages = "29--40",

editor = "\{Ali Salah\}, Albert and Javier Ruiz-del-Solar and {\c C}etin Meri{\c c}li and Pierre-Yves Oudeyer",

booktitle = "Human behavior understanding",

address = "Germany",

}

Chaaraoui, AA, Climent-Perez, P & Florez Revuelta, F 2012, An efficient approach for multi-view human action recognition based on bag-of-key-poses. in A Ali Salah, J Ruiz-del-Solar, Ç Meriçli & P-Y Oudeyer (eds), Human behavior understanding: Third Workshop, HBU 2012, Vilamoura, Portugal, October 7, 2012, proceedings. vol. 7559, Lecture Notes in Computer Science, vol. 7559, Springer, Berlin, Germany, pp. 29-40, 3rd International Workshop on Human Behavior Understanding, 7/10/12. https://doi.org/10.1007/978-3-642-34014-7_3

An efficient approach for multi-view human action recognition based on bag-of-key-poses. / Chaaraoui, Alexandros Andre; Climent-Perez, Pau; Florez Revuelta, Francisco.
Human behavior understanding: Third Workshop, HBU 2012, Vilamoura, Portugal, October 7, 2012, proceedings. ed. / Albert Ali Salah; Javier Ruiz-del-Solar; Çetin Meriçli; Pierre-Yves Oudeyer. Vol. 7559 Berlin, Germany: Springer, 2012. p. 29-40 (Lecture Notes in Computer Science; Vol. 7559).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - An efficient approach for multi-view human action recognition based on bag-of-key-poses

AU - Chaaraoui, Alexandros Andre

AU - Climent-Perez, Pau

AU - Florez Revuelta, Francisco

N1 - Organising Body: Springer

PY - 2012/10

Y1 - 2012/10

N2 - This paper presents a novel multi-view human action recognition approach based on a bag-of-key-poses. In the case of multi-view scenarios, it is especially difficult to perform accurate action recognition that still runs at an admissible recognition speed. The presented method aims to fill this gap by combining a silhouette-based pose representation with a simple, yet effective multi-view learning approach based on Model Fusion. Action classification is performed through efficient sequence matching and by the comparison of successive key poses which are evaluated on both feature similarity and match relevance. Experimentation on the MuHAVi dataset shows that the method outperforms currently available recognition rates and is exceptionally robust to actor-variance. Temporal evaluation confirms the method’s suitability for real-time recognition.

AB - This paper presents a novel multi-view human action recognition approach based on a bag-of-key-poses. In the case of multi-view scenarios, it is especially difficult to perform accurate action recognition that still runs at an admissible recognition speed. The presented method aims to fill this gap by combining a silhouette-based pose representation with a simple, yet effective multi-view learning approach based on Model Fusion. Action classification is performed through efficient sequence matching and by the comparison of successive key poses which are evaluated on both feature similarity and match relevance. Experimentation on the MuHAVi dataset shows that the method outperforms currently available recognition rates and is exceptionally robust to actor-variance. Temporal evaluation confirms the method’s suitability for real-time recognition.

KW - human action recognition

KW - multi-view action recognition

KW - key pose

KW - bag-of-key-poses

KW - MuHAVi dataset

KW - Computer science and informatics

U2 - 10.1007/978-3-642-34014-7_3

DO - 10.1007/978-3-642-34014-7_3

M3 - Conference contribution

SN - 9783642340130

VL - 7559

T3 - Lecture Notes in Computer Science

SP - 29

EP - 40

BT - Human behavior understanding

A2 - Ali Salah, Albert

A2 - Ruiz-del-Solar, Javier

A2 - Meriçli, Çetin

A2 - Oudeyer, Pierre-Yves

PB - Springer

CY - Berlin, Germany

T2 - 3rd International Workshop on Human Behavior Understanding

Y2 - 7 October 2012 through 7 October 2012

ER -

Chaaraoui AA, Climent-Perez P, Florez Revuelta F. An efficient approach for multi-view human action recognition based on bag-of-key-poses. In Ali Salah A, Ruiz-del-Solar J, Meriçli Ç, Oudeyer PY, editors, Human behavior understanding: Third Workshop, HBU 2012, Vilamoura, Portugal, October 7, 2012, proceedings. Vol. 7559. Berlin, Germany: Springer. 2012. p. 29-40. (Lecture Notes in Computer Science). doi: 10.1007/978-3-642-34014-7_3

An efficient approach for multi-view human action recognition based on bag-of-key-poses

Abstract

Publication series

Conference

Bibliographical note

Keywords

Access to Document

Fingerprint

Cite this